new panic in cpu_reset() with WITNESS

Gleb Smirnoff glebius at FreeBSD.org
Mon Jan 23 13:07:45 UTC 2012


On Mon, Jan 23, 2012 at 10:21:57AM +0200, Andriy Gapon wrote:
A> > I can confirm that I can reboot succesfully with a new kernel
A> > and kern.stop_scheduler_on_panic=0.
A> 
A> This is very puzzling.  You observation means that stop_scheduler_on_panic has
A> an effect on the code outside the panic path.  I reviewed the
A> SCHEDULER_STOPPED() changes and I still can not see how that could be possible
A> (modulo compiler bugs).
A> 
A> Unfortunately, I also can not reproduce the problem here - without
A> WITNESS_SKIPSPIN I am running into a panic[*] during boot.

I suppose your patch (the one in other post) should help against
this panic on boot.

A> BTW, do you get any other LOR reports _before_ you do a reboot?

I have some, but I don't think they are related.

A> Can you try to change printfs in witness to db_printfs?  Perhaps this will allow
A> to get the details of the LOR in uart_cnputc.  Maybe that will reveal some
A> important additional details.

Should I do s/printf/db_printf/ throughout entire subr_witness.c, or only
in special places?

A> Finally, can you try the patch from my other post?

Yes, it works.

-- 
Totus tuus, Glebius.


More information about the freebsd-current mailing list